The hypotheses for this study are: 1) Patient with Major Depressive Disorder (MDD) have a blunted GH response to the GABA-B receptor agonist baclofen, compared to healthy age-matched control and 2) The GH response to this neuroendocrine challenge will correlate with plasma GABA. This study provides evidence for whether a GABA functional deficit exists in a narrowly and carefully defined group of MDD patients. It also provides the first direct test of how well a peripheral marker of GABA reflects brain GABA function in humans.